influencer-analyzer
Pass
Audited by Gen Agent Trust Hub on Oct 31, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data retrieved from web searches and social media platforms to analyze content patterns and identify gaps for a content strategy. This creates a potential surface for indirect prompt injection, where malicious instructions embedded in influencer content could influence the agent's output.
- Ingestion points: External influencer content retrieved via Perplexity, DuckDuckGo, and web scraping as described in
SKILL.md. - Boundary markers: The prompts in
scripts/analyze_influencer.pyinterpolate context into task descriptions without explicit delimitation or 'ignore instructions' warnings for external content. - Capability inventory: The script performs API calls to Anthropic models and optionally writes Markdown reports to the local filesystem (
scripts/analyze_influencer.py). - Sanitization: The skill does not implement specific sanitization or filtering for the external content it analyzes.
Audit Metadata